Various methods have been devised for navigating robotic floor-cleaning device automatically. However, paths derived automatically are often not as efficient or desirable as paths selected by a user. A user may more quickly and easily identify areas where a robotic floor-cleaning device should not drive, for example, in an area with a lot of cords and cables or near a fragile piece of furniture. A need exists for a method to replicate a human's ability to identify an ideal path for a robotic floor-cleaning device.